Domain Specific Sentence Level Mood Extraction from Malayalam Text

• Main Authors: Mohandas, Neethu, Nair, Janardhanan P.S., V., Govindaru
• Format: Conference Proceeding
• Language: English
• Subjects: Accuracy; Learning systems; Malayalam; Mood; Natural language processing; parts-of-speech tagging; Semantic Orientation; Semantics; Sentiment analysis; Support vector machines; Tagging
• DOI: 10.1109/ICACC.2012.16

Natural Language Processing (NLP) is a field which studies the interactions between computers and natural languages. NLP is used to enable computers to attain the capability of manipulating natural languages with a level of expertise equivalent to humans. There exists a wide range of applications for NLP, of which sentiment analysis(SA) plays a major role. In sentimental analysis, the emotional polarity of a given text is analysed, and classified as positive, negative or neutral. A more difficult task is to refine the classification into different moods such as happy, sad, angry etc. Analysing a natural language for mood extraction is not at all an easy task for a computer. Even after achieving capabilities of massive amount of computation within a matter of seconds, understanding the sentiments embodied in phrases and sentences of textual information remains one of the toughest tasks for a computer till date. This paper focuses on tagging the appropriate mood in Malayalam text. Tagging is used to specify whether a sentence indicates a sad, happy or angry mood of the person involved or if the sentence contains just facts, devoid of emotions. This research work is heavily dependent on the language since the structures vary from language to language. Mood extraction and tagging has been successfully implemented for English and other European languages. For the south Indian language Malayalam, no significant work has yet been done on mood extraction. We will be focusing on domain-specific sentence-level mood extraction from Malayalam text. The extraction process involves parts-of-speech tagging of the input sentence, extracting the patterns from the input sentence which will contribute to the mood of the sentence, such as the adjective, adverb etc., and finally scoring the sentence on an emotive scale by calculating the semantic orientation of the sentence using the extracted patterns. Sentiment classification is becoming a promising topic with the rise of social media such as blogs, social networking sites, where people express their views on various topics. Mood extraction enables computers to automate the activities performed by human for making decisions based on the moods of opinions expressed in Malayalam text.